This invention relates to a method and apparatus for correcting a printhead, a printhead corrected by this apparatus, and a printing apparatus using this printhead. More particularly, the invention relates to a method and apparatus for correcting, by way of example, a full-line printhead equipped with a plurality of printing elements corresponding to the printing width of a recording medium, a printhead corrected by this apparatus, and a printing apparatus using this printhead.
A printer or the printing section of a copying machine or facsimile machine is so adapted as to print an image, which comprises a dot pattern, on a recording medium such as a paper, a thin plastic sheet or fabric based upon image information.
Among these printing apparatus, those which are the focus of attention because of their low cost are mounted with printheads that rely upon the ink-jet method, the thermosensitive-transfer method or the LED method, etc., in which a plurality of printing elements corresponding to dots are arrayed on a base.
In a printhead in which these printing elements are arrayed to correspond to a certain printing width, the printing elements can be formed through a process similar to a semiconductor manufacturing process. Accordingly, a transition is now being made from a configuration in which the printhead and driving integrated circuitry are arranged separately of each other to an integrated assembled configuration in which the driving integrated circuitry is structurally integrated within the same base on which the printing elements are arrayed. As a result, complicated circuitry involved in driving the printhead can be avoided and the printing apparatus can be reduced in size and cost.
Among these types of printing methods, the ink-jet printing method is particularly advantageous. Specifically, according to this method, thermal energy is made to act upon ink and the ink is discharged by utilizing the pressure produced by thermal expansion. This method is advantageous in that the response to a printing signal is good and it is easy to group the orifices close together at a high density. This method attracts a great deal of attention in comparison with the other methods.
When the printhead is manufactured by applying a semiconductor manufacturing process and, in particular, when numerous printing elements that are to be made to correspond to the printing width are arrayed over the entire area of a base, it is very difficult to manufacture all of the printing elements without any defects. As a consequence, the manufacturing yield of the process for manufacturing the printhead is poor and is accompanied by higher cost. There are occasions where such a printhead cannot be put into practical use because of the costs involved.
Accordingly, methods of obtaining a full-line printhead have been disclosed in the specifications of Japanese Patent Application Laid-Open (KOKAI) Nos. 55-132253, 2-2009, 4-229278, 4-232749 and 5-24192 and in the specification of U.S. Pat. No. 5,016,023. According to these methods, a number of high-yield printheads each having an array of printing elements of a comparatively small number of orifices, e.g., 32, 48, 64 or 128 printing elements, are placed upon (or upon/below) a single base at a high precision in conformity with the density of the array of printing elements, thereby providing a full-line printhead whose length corresponds to the necessary printing width.
It has recently become possible on the basis of this technique to simply manufacture a full-line printhead by arraying printing elements of a comparatively small number (e.g., 64 or 128) of orifices on bases (also referred to as "printing units") and bonding these printing units in a row on a base plate in highly precise fashion over a length corresponding to the necessary printing width.
Though it has thus become easy to manufacture a full-line printhead, certain performance-related problems remain with regard to a printhead manufactured by the foregoing manufacturing method. For example, a decline in printing quality, such as density unevenness, cannot be avoided. The cause is a variation in performance from one printing unit (base) to another in the row of such printing units, a variation in the performance of neighboring printing elements between the arrayed printing units and heat retained in each driving block at the time of recording.
In particular, in the case of an ink-jet printhead, not only a variation in the neighboring printing elements between the arrayed printing units but also a decline in ink fluidity owing to the gaps between printing units results in lower yield in the final stage of the printhead manufacturing process. For this reason, the state of the art is such that these printheads are not readily available on the market in large quantities regardless of the fact these printheads exhibit highly satisfactory capabilities.
As means for correcting density unevenness caused by the printhead, Japanese Patent Application No. 6-34558 discloses a method of correcting the unevenness in the density of a printhead by measuring dot diameter and correcting unevenness based upon the results of measurement. However, there is a room for improving reproducibility of printed dots. For example, when one line of printing has been performed, the characteristics of the printed dots change subtly on the next line, over then next several dozen lines and over the next several hundred lines. (This is known as "fluctuation" from dot to dot.) However, since a specific phenomenon (dot diameter) which incorporates this fluctuation is conventionally employed as information regarding density unevenness, it is difficult to obtain satisfactory results with a single correction. In order to acquire the desired image quality, it is required that printed dot data from several measurements be acquired to perform the correction. In a case where electrical energy is converted to thermal energy in conformity with correction data, energy which is larger than usual is applied to the printing elements that exhibit a low density. Thus, there is still a room for further improving reliability in terms of the durability of the printhead.
Furthermore, it has been sometimes impossible to assuredly correct the density unevenness according to a method for predicting on the basis of OD values, which is one of conventional density unevenness correction methods, since there has been some unevenness in manufacturing each of printheads and the method does not show a good correlation.
Also, if one reference OD value is selected on the basis of a phenomenon due to the combinations of driving control pulses for n times, the density unevenness can be corrected when the range of the density unevenness is small, however, there has been a limitation for correcting the density unevenness over the wide range of variation of density.